Relevance of Endocrine Pancreas Nesidioblastosis To
Total Page:16
File Type:pdf, Size:1020Kb
EDITORIALS plasia) rather than a continuous proliferation and differ- Relevance of Endocrine entiation of endocrine cells (19-21). Pancreas Nesidioblastosis to Note that normoglycemia can occur in various indi- viduals having a wide disparity in endocrine cell mass Hyperinsulinemic Hypoglycemia (20). Therefore, it is difficult to reconcile how a small Whereas surgery often concludes the saga of hypo- increase in p-cell mass could cause hyperinsulinemic glycemia, the pathologist has the final word. In children hypoglycemia, assuming the (3-cells are functionally and occasionally in adults, that word usually is nesidio- normal. The frequent recurrence of hypoglycemia after blastosis, a somewhat ill-defined and nebulous diag- 60-80% pancreatic resection also suggests that a factor nosis (1-6). Exactly what is nesidioblastosis? Does it other than increased (3-cell mass is involved. represent any sort of pathology at all? Because somatostatin has been shown to inhibit in- More than 30 years after the initial description of in- sulin secretion, a quantitative deficiency of 8-cells (or a fantile hypoglycemia by McQuarric (7), the pathogen- loss of cellular contacts between (S- and 8-cells) has esis of hyperinsulinism remains unclear in most cases. been incriminated as the etiology of the disease (22- Few resected glands from these hypoglycemic infants 25). Several studies have demonstrated a reduced den- show focal lesions: they either consist of a true adenoma sity of 8-cells in hypoglycemic infants. However, this exhibiting a compact ribbonlike pattern reminiscent of reduction is not present in all infants and must be in- that observed in islet cell tumors of adults or of focal terpreted with caution, because degranulation of 8-cells adenomatous hyperplasia. Surgical resection limited to may impair immunohistochemical detection. An appar- the lesion is usually followed by disappearance of the ently normal topographic relationship between insulin hypoglycemia, which indicates that the focal abnor- and somatostatin cells in these cases has been reported, mality was the cause of the hyperinsulinism. but abnormalities in the intercellular communications In most cases, however, no focal lesion is detectable. have not been excluded (6). Instead, a diffuse and disseminated proliferation of islet Several groups have observed abnormally large 0-cell cells budding off from pancreatic ducts is usually ob- nuclei in these infants (8,19,26-31). This observation served and has been proposed repeatedly as the under- also suggests the existence of functional abnormalities, lying pathological lesion (8-12,14-17). The term because the size of the nucleus may reflect the func- nesidioblastosis, currently used to describe this condi- tional activity of endocrine cells (32). In a series of 16 tion, was originally coined by Laidlaw (18) because infants with hyperinsulinism and hypoglycemia, such an pancreatic endocrine tumors originate from nesidio- increase of fJ-cell nuclear volume was demonstrated by blasts, the stem cells that differentiate from the duct. morphometry in all cases without focal lesions. Indeed, Subsequently, Yakovac (9) defined nesidioblastosis as a increased (3-cell nuclear volume may constitute a mor- continuous differentiation of (3-cells from the ductular phologic criterion useful in differentiating focal from dif- system of the exocrine pancreas. fuse lesions; although similarly large nuclei have been Several recent studies that used quantitative immu- observed in adenomas, they appear in such cases to be nohistochemical techniques have demonstrated that restricted to the focal lesion (30). Hypertrophic (3-nuclei endocrine cells in close contact with ducts (nesidio- are also present in the pancreas glands of infants born blastosis) can also be observed in normoglycemic to diabetic mothers. Whether these nuclear features ob- infants and young children and thus are not pathogno- served in hyperinsulinemic infants reflect a primary ab- monic for the disease (15-21). Unfortunately, nesidio- normality of the p-cells or whether they are secondary blastosis is a morphologic parameter difficult to quan- to a loss or an impairment of the mechanisms controlling titate, particularly in the neonatal and infantile pancreas (3-cell function remains to be established. in which scattered endocrine cells or small clusters are In this issue, Fong et al. (p. 108) remind us that hy- numerous. Even when careful morphometric studies perinsulinemic hypoglycemia without endocrine tumor have been performed, it cannot be certain that these also occurs occasionally in adults (1-6). In adults and morphologic findings are more prominent in hypogly- children, nesidioblastosis repeatedly has been proposed cemic than in normoglycemic infants. On the other as the morphologic abnormality responsible for the hy- hand, precise quantitative studies have established that perinsulinism. However, nesidioblastosis has been ob- the total pancreatic mass of endocrine cells is not in- served in normoglycemic adults in whom the frequency creased in hyperinsulinemic infants (30). Furthermore, of ductuloinsular complexes has been quantified (33). more numerous small endocrine clusters in hypo- Similarly, nesidioblastosis has been reported in associ- glycemic than normoglycemic infants might reflect an ation with several clinical conditions without hyperin- abnormality in the formation of true islets (nesidiodys- sulinemic hypoglycemia (e.g., cystic fibrosis, chronic pancreatitis, overweight, endogenous hypergastrinemia, congenital heart malformation, deficiency of a,-protein- ase inhibitor, and androgen therapy; (34-40). There- From the Department of Pathology, University Hospital St. Luc, Brussels, Bel- fore, it appears that the presence of endocrine cells gium. budding off from ducts or of small clusters of p-cells Address correspondence and reprint requests to Jacques Rahier, MD, Uni- versity Hospital St. Luc, Avenue Hippocrate 10, B1200 Brussels, Belgium. scattered throughout the exocrine pancreas is not suffi- 164 DIABETES CARE, VOL. 12, NO. 2, FEBRUARY 1989 EDITORIALS cient to explain excessive insulin release. In certain con- 12. Thomas CG, Underwood LE, Carney CN, Dolcourt JL, ditions, nesidioblastosis could reflect (3-cell replication Whitt JJ: Neonatal and infantile hypoglycemia due to in- compensating for (3-cell destruction (chronic pancrea- sulin excess: new aspects of diagnosis and surgical man- titis) or functional insufficiency. In such cases, it might agement. Ann Surg 185:505-17, 1977 14. Heitz PU, Kloppel G, Hacki WH, Polak JM, Pearse AH: represent a compensatory mechanism for normalizing Nesidioblastosis: the pathologic basis of persistent hyper- the functional (3-cell mass. Note that nesidioblastosis insulinemic hypoglycemia in infants: morphologic and has also been reported in a patient with hyperinsulin- quantitative analysis of seven cases based on specific im- emic hypoglycemia caused by ingestion of chlorpro- munostaining and electron microscopy. Diabetes 26: pamide, although whether nesidioblastosis represented 632-42, 1977 an associated or secondary phenomenon in this case is 15. Becker K, Wendel U, Przyrembel H, Tsotsalas M, Mun- unknown (41). tefering H, Bremer HJ: Beta cell nesidioblastosis. Eur } Finally, nesidioblastosis with hyperinsulinism might Pediatr 127:75-89, 1978 reflect a secondary response to some unrecognized 16. Dahms BB, Landing BH, Blaskovics M, Roe TF: Nesid- stimulating factor or, conversely, a peculiar responsive- ioblastosis and other islet cell abnormalities in hyperin- sulinemic hypoglycemia of childhood. Hum Pathol 11: ness of the endocrine pancreas to normal stimuli. 641-49, 1980 The coexistence of hyperinsulinemic hypoglycemia 17. Aynsley-Green A: Nesidioblastosis of the pancreas in in- and nesidioblastosis does not prove that they are caus- fancy. In Carbohydrate Metabolism and Its Disorders. ally related. Because human pancreatic kinetic studies Randle PJ, Steiner DF, Whelan WJ, Eds. London, Aca- are still lacking, whether nesidioblastosis reflects accel- demic, 1981, p. 181-204 erated endocrine cell replication is even uncertain. 18. Laidlaw GF: Nesidioblastoma, the islet tumor of the pan- creas. Am ) Pathol 14:125-34, 1938 JACQUES RAHIER, RD, MD, PhD 19. Jaffe R, Hashida Y, Yunis EJ: Pancreatic pathology in hy- perinsulinemic hypoglycemia of infancy. Lab Invest 42: 356-65, 1980 20. Rahier J, Wallon J, Henquin JC: Cell populations in the REFERENCES endocrine pancreas of human neonates and infants. Dia- betologia 20:540-46, 1980 1. Harness JK, Ceelhoed CW, Thompson NW, Nishiyama 21. Gould VE, Memoli VA, Dardi LE, Gould NS: Nesidio- RH, Fajans SS, Kraft RO, Howard DR, Clark KA: Nesidio- dysplasia and nesidioblastosis of infancy. Scand J Gas- blastosis in adults: a surgical dilemma. Arch Surg 116: troenterol 16:129-42, 1981 575-80, 1981 22. Heitz PU, Kloppel G, Polak JM: Morphology of the en- 2. Keller A, Stone AM, Valderrama E, Kolodny H: Pancreatic docrine pancreas in persistent hypoglycemia in infants. nesidioblastosis in adults: report of a patient with hyper- In Current Views on Hypoglycemia and Clucagon. An- insulinemic hypoglycemia. Am I Surg 145:412-16, 1981 dreani D, Lefevre PJ, Marks V, Eds. London, Academic, 3. Nathan DM, Axelrod L, Proppe KH, Wald R, Hirsch HJ, 1980, p. 355-65 (Proc. Serono Symp. 30) Martin DB: Nesidioblastosis associated with insulin-me- 23. Bishop AE, Polak JM, Chesa PG, Timson CM, Bryant MG, diated hypoglycemia in an adult. Diabetes Care 4:383- Bloom SR: Decrease of pancreatic somatostatin